Dr. G. C. Wallich on the Diatomacea. 19 



stipes and cushion-like masses of the pedunculate and filamcn- 

 tous forms to be merely modifications of the filament in one or 

 both directions ; and it is highly probable, therefore, that the 

 remarkable unsymTnetrically arranged processes referred to by 

 me (Quart. Journ. Micr. Science, vol. vi. p. 423) as occurring in 

 both species of the genus Hydrosera, and also to be found in 

 several Indian stipitate forms, are in reality apertures, through 

 which the modified filament is extended. Of course, in the 

 discoid forms, the marginal processes would perform the same 

 office*. 



In Bacillaria paradoxa we have to assume the presence of a 

 highly elastic envelope, embracing the entire filament, to enable 

 us to conceive how the several series of prehensile filaments may 

 produce the curious movement which that form exhibits. That 

 such envelopes are to be found both amongst the DiatomaceaB 

 and Desmidiacese, is well known. And in proof of the elastic 

 character which analogous structures may possess, it is only 

 necessary for me to state that, from observations made by me 

 on an Indian species of Schizonema, the character spoken of was 

 remarkably seen in the enveloping gelatinous sheath, each frus- 

 tule, after passing along in continuous and regular order till it 

 reached the torn orifice of the sheath, being there suddenly and 

 very forcibly expelled to some distance. 



I may, in conclusion, add that every effort to render visible 

 the filaments whose existence I assume, either by iodine or re- 

 agents, has hitherto failed. In the endosmotic and exosmotic 

 theory of Professor Smith, it became necessary to assume not 

 only as much, but more, the evidence of currents., which were 

 essential to its truth, being deficient. But it is surely unphilo- 

 sophical to deny the existence of all we are unable to see. We 

 may still hit upon some vegetable colouring or dyeing matter, 

 of innocuous quality to the Diatom, but capable of staining its 

 most delicate portions. This innocuous character is essential, 

 inasmuch as in any attempt to detect the nature of highly deli- 

 cate organs, such as the most attenuated ciliary apparatus or 



* A novel form ofCoscinodiscus (provisionally named by me C. Sol) is found 

 in the stomach-contents of Salpce, It has a broad membranous plate extend- 

 ing round the periphery of the silicious disk, across which are arranged, 

 radially, very delicate folds springing from the margin of the silicious por- 

 tion. This membranous plate may be looked upon as another modifica- 

 tion of the filamentous organs. The occurrence of so very remarkable an 

 appendage to a Diatom affords a valuable confirmation of the view given 

 'above. 



It may be mentioned that, on exposure to acid, the membranous plate 

 is dissolved, but the silicious disk remains and is then un distinguishable 

 from an ordinary C. radiatus of small size. Specimens in my possession 

 exhibit the two valves and plates in all their integrity. 



2* 



